From mboxrd@z Thu Jan 1 00:00:00 1970 From: Maxime Coquelin Subject: Re: [PATCH v2 3/4] clk: stm32: Add clock driver for STM32F4[23]xxx devices Date: Sat, 30 May 2015 11:15:58 +0200 Message-ID: <55697FCE.9040003@gmail.com> References: <1432327273-6810-1-git-send-email-daniel.thompson@linaro.org> <1432972448-10332-1-git-send-email-daniel.thompson@linaro.org> <1432972448-10332-4-git-send-email-daniel.thompson@linaro.org> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1432972448-10332-4-git-send-email-daniel.thompson@linaro.org> Sender: linux-kernel-owner@vger.kernel.org To: Daniel Thompson , Mike Turquette , Stephen Boyd Cc: Rob Herring , Pawel Moll , Mark Rutland , Ian Campbell , Kumar Gala , Russell King , Kamil Lulko , Andreas Farber , linux-clk@vger.kernel.org, devicetree@vger.kernel.org, linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org, patches@linaro.org, linaro-kernel@lists.linaro.org List-Id: devicetree@vger.kernel.org Hi Daniel, Nice driver, please find my comments below. Once fixed, you can add: Reviewed-by: Maxime Coquelin On 05/30/2015 09:54 AM, Daniel Thompson wrote: > The driver supports decoding and statically modelling PLL state (i.e. > we inherit state from bootloader) and provides support for all > peripherals that support simple one-bit gated clocks. The covers all > peripherals whose clocks come from the AHB, APB1 or APB2 buses. > > It has been tested (for non-regression only) on an STM32F429I-Discovery > boards. The clock counts for TIM2, USART1 and SYSTICK are all set correctly > and time and the wall clock looks OK when checked with a stopwatch. > > Signed-off-by: Daniel Thompson > --- > drivers/clk/Makefile | 1 + > drivers/clk/clk-stm32f4.c | 364 ++++++++++++++++++++++++++++++++++++++++++++++ > 2 files changed, 365 insertions(+) > create mode 100644 drivers/clk/clk-stm32f4.c ... > diff --git a/drivers/clk/clk-stm32f4.c b/drivers/clk/clk-stm32f4.c > new file mode 100644 > index 0000000..68f9962 > --- /dev/null > +++ b/drivers/clk/clk-stm32f4.c > @@ -0,0 +1,364 @@ ... > +/* > + * Converts the primary and secondary indices (as they appear in DT) to an > + * offset into our struct clock array. > + */ > +static unsigned int stm32f4_rcc_lookup_clk_idx(u8 primary, u8 secondary) > +{ > + u64 table[ARRAY_SIZE(stm32f42xx_gate_map)]; > + > + if (primary == 1) { > + BUG_ON(secondary > FCLK); Maybe the function could return a signed int, an propagate errors instead of using BUG_ON? > + return secondary; > + } > + > + memcpy(table, stm32f42xx_gate_map, sizeof(table)); > + > + /* only bits set in table can be used as indices */ > + BUG_ON(secondary > 8 * sizeof(table) || > + 0 == (table[BIT_ULL_WORD(secondary)] & BIT_ULL_MASK(secondary))); Ditto. > + > + /* mask out bits above our current index */ > + table[BIT_ULL_WORD(secondary)] &= > + GENMASK_ULL(secondary % BITS_PER_LONG_LONG, 0); > + > + return FCLK + hweight64(table[0]) + > + (BIT_ULL_WORD(secondary) >= 1 ? hweight64(table[1]) : 0) + > + (BIT_ULL_WORD(secondary) >= 2 ? hweight64(table[2]) : 0); > +} > + > +struct clk *stm32f4_rcc_lookup_clk(struct of_phandle_args *clkspec, void *data) > +{ > + return clks[stm32f4_rcc_lookup_clk_idx(clkspec->args[0], > + clkspec->args[1])]; If stm32f4_rcc_lookup_clk_idx() returns an error, you could propagate it using ERR_PTR(). > +} > + > +static const char __initdata *sys_parents[] = { "hsi", NULL, "pll" }; > + > +static struct clk_div_table ahb_div_table[] = { Should be const. > + { 0x0, 1 }, { 0x1, 1 }, { 0x2, 1 }, { 0x3, 1 }, > + { 0x4, 1 }, { 0x5, 1 }, { 0x6, 1 }, { 0x7, 1 }, > + { 0x8, 2 }, { 0x9, 4 }, { 0xa, 8 }, { 0xb, 16 }, > + { 0xc, 64 }, { 0xd, 128 }, { 0xe, 256 }, { 0xf, 512 }, > + { 0 }, > +}; > + > +static struct clk_div_table apb_div_table[] = { Ditto. > + { 0, 1 }, { 0, 1 }, { 0, 1 }, { 0, 1 }, > + { 4, 2 }, { 5, 4 }, { 6, 8 }, { 7, 16 }, > + { 0 }, > +}; > Thanks! Maxime